Metatranscriptomics holds the prospect of predicting fungal phenotypes based on patterns of gene expressions, providing new opportunities to obtain information about metabolic processes without disturbance of natural systems, and with taxonomic resolution. Acquisition of fungal metabolic carbon and its subsequent partitioning between biomass production and respiration, i.e. the carbon-use efficiency, are central parameters in biogeochemical modelling. However, current available techniques for estimating these parameters in natural systems are all associated with practical and theoretical shortcomings, making assessments unreliable. We cultured four different fungal isolates (Chalara longipes, Laccaria bicolor, Serpula lacrymans, and Trichoderma harzianum) in liquid media with contrasting nitrogen availability and measured growth rates and respiration to calculate carbon-use efficiency. By relating expression of gene markers to measured carbon fluxes, we identified genes coding for 1,3-β-glucan synthase and 2-oxoglutarate dehydrogenase as good markers for growth and respiration, respectively, capturing both intraspecific variation as well as within-strain variation dependent on growth medium. A gene expression index based on these markers correlated significantly with differences in carbon-use efficiency between the fungal isolates. Our study paves the way for use of these markers to assess differences in growth, respiration, and carbon-use efficiency in natural fungal communities, using metatranscriptomic or RT-qPCR approach.

Culture performance

Glucose concentration in the medium was measured at regular intervals using a GM-100 glucose monitoring system (BioReactor Sciences, Lawrenceville, GA, USA). The systems were harvested when roughly 30% of the glucose in the medium had been consumed (Figure S1; Table 1). Immediately before harvest, respiration rates were measured using an EGM-4 portable infra-red gas analyser (PP Systems, Amesbury, MA, USA) with a closed sampling loop that incorporated the fungal culture bottles via the sterile filters. Prior to measurement, the systems were flushed with filtered air for 1 hour to allow dissolved CO₂ to equilibrate with the atmosphere. CO₂ accumulation was measured during 2.5 minutes for each system.

RNA extraction, sequencing, and bioinformatic analyses

Mycelium was harvested by filtration through Whatman™ filter paper; ø 55m, pore size 12 mm (ThermoFisher Scientific, Waltham, MA, USA) and immediately shock-frozen with liquid nitrogen, freeze dried, weighted using an ES120A analytical balance (Precisa Gravimetrics, Dietikon, Switzerland) and stored at -80°C for mRNA extraction. Total RNA was extracted from the harvested mycelium using the RNA mini kit (Qiagen, Hilden, Germany) and cleaned from remaining DNA by the DNAse I kit (Sigma-Aldrich®, St. Louis, MO, USA). Poly-A selection and mRNA library preparation was conducted using the TruSeq library preparation kit (Illumina, San Diego, CA, USA). Libraries were sequenced on the Illumina NovaSeq 6000 SP platform, yielding 50 bp paired-end sequences. Poly-A selection of the mRNA, library preparation, and sequencing were performed by the SNP&SEQ Technology Platform of SciLifeLab, Uppsala, Sweden.

Raw paired-end reads were subjected to quality control using FastQC (Andrews et al. 2010). Sequencing adapter trimming and removal of low quality bases were performed in the program ‘Trimmomatic’ (Bolger, Lohse and Usadel 2014) using the default settings. Reference genomes and gene annotations of T. harzianum (Druzhinina et al. 2018), C. longipes (Barbi et al. 2020), L. bicolor (Martin et al. 2008), and S. lacrymans (Eastwood et al. 2011) were retrieved from the JGI—Mycocosm database (https://mycocosm.jgi.doe.gov/mycocosm/home). Filtered mRNA sequences were mapped against respective genomes using ‘bowtie2’ (Langmead and Salzberg 2012). Data was sorted, indexed, and converted to gene expression count tables using ‘SAMtools’ (Li et al. 2009; Li 2011).  Gene expression data were normalized for gene lengths and sequencing effort according to the RPKM method (reads per kilo base per million mapped reads) (Mortazavi et al. 2008). To enable analysis of expression of enzyme-encoding genes across fungal isolates, data was aggregated according to Enzyme Commission (EC) numbers, which denote a numerical classification of enzymes (Kanehisa 2017).

This dataset consists of 2 files containing information about culture performance (carbon uptake, growth, respiration etc.), expression of gene markers used (RPKM normalised), and transcriptome (RPKM normalised and filtered). Missing values are indicated by 'NA'.